Cytoplasmic conductivity in apoptotic CHO cells falls from roughly 0.45 S/m in viable cells to approximately 0.05 S/m — a ninefold drop that label-free dielectrophoresis (DEP) can now capture within seconds, according to a new review covering advances from 2015 through 2026.

The review, published in Electrophoresis (DOI: 10.1002/elps.70146), was written by Alaleh Vaghef-Koodehi, PhD, a postdoctoral researcher at the University of Massachusetts Amherst, and Blanca Lapizco-Encinas, PhD, professor of biomedical engineering at the Rochester Institute of Technology. It examines how membrane capacitance, cytoplasmic conductivity, polarizability and surface charge can reveal cellular state without labels — and what that means for bioprocess control.

Why does timing matter in biologics manufacturing?

The authors call real-time monitoring of cell health a "critical bottleneck" in biologics production. Conventional apoptosis assays — offline fluorescent staining, lactate dehydrogenase release — are destructive and offer only limited temporal resolution. By the time they detect trouble, options for the batch may already be narrow.

DEP offers a different entry point. The technique exploits how polarizable cells respond to nonuniform electric fields, tracking changes in their electrical signature continuously, without labeling or destroying the sample. For a process development team weighing at-line versus offline analytics, that distinction is the core of the argument.

What did the CHO cell studies actually measure?

Chinese hamster ovary cells — the review's "industry workhorse" — supplied the sharpest data points. Studies using dual-frequency DEP cytometry followed CHO cultures through nutrient-starvation-induced apoptosis and identified two distinct electrical phases:

• Membrane remodeling first, with a gradual reduction in membrane capacitance
• A subsequent sharp drop in cytoplasmic conductivity as ionic homeostasis broke down

In one study, apoptotic populations began appearing between 24 and 36 hours, with cytoplasmic conductivity falling to approximately 0.05 S/m versus about 0.45 S/m in viable cells. A follow-up study found membrane capacitance declining gradually before cytoplasmic conductivity plunged to 0.07 S/m in apoptotic cells by 52 hours.

Together, the two phases provide an electrical timeline of cell death rather than a single endpoint measurement. That is the practical difference for harvest planning: a trajectory instead of a verdict.

How fast is fast enough for the manufacturing floor?

Speed is the platform's strongest selling point. The commercially available 3DEP platform has captured electrical signatures from populations of roughly 20,000 cells within seconds. According to the review, such measurements can reveal emerging apoptotic subpopulations "hours before traditional biochemical markers become detectable."

For operators, that lead time translates into decisions: adjust feeding or process parameters, or pull the harvest forward before culture performance deteriorates. The measured capability is detection hours earlier; the downstream gains in yield or batch salvage remain projections that biomanufacturers would need to validate on their own processes.

Can measurements become automated decisions?

The review points to a 2026 analysis that integrated electrical impedance data with supervised machine learning. A tuned random-forest model reached 90% predictive accuracy and was proposed as a label-free "traffic-light" system for batch monitoring.

That number comes from a single model on a defined dataset, so treat it as a proof of concept rather than a validated control strategy. Still, the trajectory the authors describe — rapid electrical measurement combined with computational analysis — points toward autonomous intervention and optimized harvesting.

What is realistic today?

Commercial DEP platforms are already increasing accessibility, moving the technique beyond specialized laboratory experiments. The authors see continued integration with computational modeling and machine learning as the route toward real-time automated systems.

Routine autonomous bioprocess control is not here yet. But the review's case is concrete: the electrical life of a cell is becoming another process signal, one that can warn a culture is changing before conventional assays catch up.